Posted 21 May, 2026
Senior Machine Learning Engineer
Xebia
Bengaluru, KA, IN
Full Time
Reference: f216bd7f1c170057
Job Description
Job Title : Senior Machine Learning Engineer\nJob Location : Bengaluru\nExp : 6-12 years\n\nPosition Overview : We are looking for a Machine learning Engineers with strong expertise on ML /AI . Should have good experience working on ML models developments and deploying the ML platforms\n\nRoles & Responsibilities :\n\nDesign, build, and optimize scalable batch and streaming data pipelines using distributed framework like Spark/Databricks/Kafka.\nDesign & Code - robust data models, feature pipelines, and ETL/ELT frameworks for analytics and ML.\nEnsure data quality, observability, lineage, and performance across data platforms.\nBuild and refine ML models end‑to‑end: feature engineering, training, evaluation, and deployment.\nPartner with data scientists to convert prototypes into production‑grade ML solutions.\nImplement CI/CD, model versioning, monitoring, and automation across data and ML workflows.\nProduct Driven Mindset: Collaborate with engineering, product teams to deliver data‑driven outcomes.\n\nSkill sets Required :\n\n6+ years of experience in ML-Data Engineering development.\nStrong SQ/NoSQL, Python, PySpark, and ML Models Lifecycle & Frameworks (MlFlow, Spark-ml), Orchestration (Airflow/Oozie/Dagster etc)\nExpertise in Big Data modeling, Distributed processing, and Lake & Warehouse architectures at large operational scale.\nHands‑on with ML lifecycle tools (MLflow, Feature Store, model monitoring, Evaluation).\nStrong Analytical & Problem Solving Skills - Data/Process Intensive Design/Architecture, Strong debugging, optimization.\nBasic hold on foundational modelling concepts & algorithms such as - Regression, Classification and Statistical models.\nGood Hold on Concepts- Distributed File Formats, Open table Formats, Distributed transaction management, Workload Parallelizing.\nJands on - Unix, Hadoop, Object store fundamental operations & commands\nBasic skilled with containerized processing (Docker + K8s)